asterixdb-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF subversion and git services (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (ASTERIXDB-2195) Support Dynamic Replica Placement
Date Mon, 08 Jan 2018 18:05:00 GMT

    [ https://issues.apache.org/jira/browse/ASTERIXDB-2195?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16316687#comment-16316687
] 

ASF subversion and git services commented on ASTERIXDB-2195:
------------------------------------------------------------

Commit 3316bde186e6447c05a0aa1afcf5c82e5df83f4f in asterixdb's branch refs/heads/master from
[~mhubail]
[ https://git-wip-us.apache.org/repos/asf?p=asterixdb.git;h=3316bde ]

[ASTERIXDB-2195][REPL] Clean Masked Files

- user model changes: no
- storage format changes: no
- interface changes: no

Details:
- Clean invalid masked files before promoting
  a partition or sending partition files list
  to master.
- Let replica calculate component id instead
  of sending it from master.
- Add tests for:
  - Deleting masked component.
  - Deleting masked file.

Change-Id: Ib0f0159159faf87b9f5fd2eca3956dd90633bcfa
Reviewed-on: https://asterix-gerrit.ics.uci.edu/2268
Sonar-Qube: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Integration-Tests: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Tested-by: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Contrib: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Reviewed-by: Michael Blow <mblow@apache.org>


> Support Dynamic Replica Placement
> ---------------------------------
>
>                 Key: ASTERIXDB-2195
>                 URL: https://issues.apache.org/jira/browse/ASTERIXDB-2195
>             Project: Apache AsterixDB
>          Issue Type: Improvement
>          Components: REPL - Replication
>            Reporter: Murtadha Hubail
>            Assignee: Murtadha Hubail
>
> In the current implementation of replication, each node has to know its replicas in advance
and establish communications with them on startup. While this works for a cluster with a fixed
topology, to support clusters with dynamic topologies, we need to introduce APIs that allow
a replica to be added at runtime. This should eventually replace the existing static fixed
fault tolerance strategies for metadata node and chained declustering and gives the cluster
admin the ability to define any replication/fault tolerance pattern.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message